A micro-environment monitoring method and device for a photovoltaic power station, an electronic device and a medium

Abstract

This application provides a method, device, electronic equipment, and medium for monitoring the microenvironment of a photovoltaic (PV) power station. The PV power station is divided into an under-panel area, an inter-panel area, and an external array area. Different types of sensor groups are deployed in each area. Initial microenvironmental parameters of the PV power station are acquired from the sensor groups in each area. Using the initial microenvironmental parameters of the external array area as constraints, a preset correction algorithm is used to correct the initial microenvironmental parameters of the under-panel and inter-panel areas, yielding the final microenvironmental parameters. This invention achieves logical verification of the complex internal microenvironment by introducing macroscopic data from outside the array as constraints. It effectively eliminates false high-temperature or high-wind readings caused by localized strong radiation, dust accumulation, or sensor drift, significantly improving the accuracy and robustness of the monitoring data.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation technology, and more specifically, to a method, device, electronic equipment, and medium for monitoring the microenvironment of a photovoltaic power station. Background Technology

[0002] Existing photovoltaic (PV) power plant monitoring technologies have significant limitations. For example, in large ground-mounted power plants, a standard weather station (which only collects ambient temperature, wind speed, and irradiance) is often set up only at the outer edge of the array. This leads to measurement errors when the sensors inside the PV array are affected by localized strong thermal radiation, dust accumulation, or airflow turbulence, resulting in drift. This reduces the accuracy and reliability of the monitoring data, making it difficult to meet the needs of refined operation and maintenance and efficient power generation in PV power plants. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method, device, electronic device and medium for monitoring the microenvironment of photovoltaic power plants, so as to improve the accuracy of monitoring environmental parameters of photovoltaic power plants.

[0004] Firstly, a method for monitoring the microenvironment of a photovoltaic (PV) power station is provided. The PV power station is divided into an under-panel area, an inter-panel area, and an external array area; different types of sensor groups are deployed in each area; the method includes: Acquire the initial microenvironmental parameters of the photovoltaic power station collected by sensor groups in each region; Using the initial microenvironment parameters of the outer region of the array as constraints, the initial microenvironment parameters of the under-board region and the inter-board region are corrected using a preset correction algorithm to obtain the final microenvironment parameters.

[0005] Optionally, before correcting the initial microenvironment parameters, the method further includes: Outlier removal was performed on the initial microenvironment parameters of each region to obtain the first data. The first data is completed using linear interpolation to obtain the second data; The second data is normalized and mapped to a unified interval to obtain the preprocessed initial microenvironment parameters.

[0006] Optionally, the initial microenvironment parameters of the region outside the array are used as constraints, and a preset correction algorithm is used to correct the initial microenvironment parameters of the under-board region and the inter-board region to obtain the final microenvironment parameters, including: The initial microenvironment parameters of each region are internally fused to obtain single fused data for each region. Using the single fused data of the outer region of the array as a priori, the Bayesian estimation method is used to correct the spatial error of the single fused data of the under-panel and inter-panel regions to obtain the intermediate microenvironment parameters of the three regions. Based on the Kalman filter algorithm, the dynamic perturbation of the intermediate microenvironment parameters over time is corrected to obtain the final microenvironment parameters.

[0007] Optionally, based on the Kalman filter algorithm, the dynamic perturbations of the intermediate microenvironment parameters over time are corrected to obtain the final microenvironment parameters, including: Using the intermediate microenvironment parameters of the previous moment as input, the predicted parameters of the current moment are calculated based on the pre-built state-space model; Based on the deviation between the predicted parameters and the actual observation parameters of the sensor group at the current moment, the predicted parameters are corrected to obtain the final microenvironment parameters.

[0008] Optionally, the process of constructing the state-space model includes: Based on the laws of energy conservation, mass conservation, and momentum conservation, a set of global heat flow coupled evolution equations characterizing regional heat and mass transfer and fluid motion is established. The photovoltaic array is divided into several grid cells. The global heat flow coupling evolution equations are discretized on the grid cells using the finite volume method to obtain the state evolution equations of each grid cell. The state evolution equations are analyzed to extract the thermal coupling coefficient, momentum transfer operator, and mass transfer operator that characterize the interaction between adjacent units; The thermal coupling coefficient, momentum transfer operator, and mass transfer operator are mapped to their corresponding positions in the preset state transition matrix framework to generate the state transition matrix, and the state space model is constructed using the state transition matrix.

[0009] Optionally, sensor groups are deployed in each area according to a preset sensor layout scheme. The process of determining the preset sensor layout scheme includes: Obtain initial rough data of the microenvironment of the photovoltaic power plant site; Spatial variation analysis was performed on the initial coarse-collected microenvironment data to extract spatially relevant distance parameters; Based on the pre-defined correspondence between spatially relevant distance parameters and the degree of parameter variation, homogeneous regions and heterogeneous mutation zones of parameter variation are identified. Based on the identified homogeneous regions and heterogeneous transition zones, the sensor deployment density in each region is adjusted to obtain an adaptive sensor layout scheme for each region.

[0010] Optionally, the method also includes: During the operation of the photovoltaic power station, the variance of the microenvironment parameters in each zone is monitored in real time. If the variance of any region exceeds a preset variance threshold for a preset number of consecutive preset times, the sensor layout determination process will be retried, and suggestions for sensor movement or addition will be generated.

[0011] Secondly, a microenvironment monitoring device for a photovoltaic power station is provided. The photovoltaic power station is divided into an under-panel area, an inter-panel area, and an external array area; different types of sensor groups are deployed in each area; the device includes: The acquisition unit is used to acquire the initial microenvironmental parameters of the photovoltaic power station collected by the sensor groups in each area; The calibration unit is used to use the initial microenvironment parameters of the outer region of the array as constraints, and to use a preset calibration algorithm to calibrate the initial microenvironment parameters of the under-board region and the inter-board region to obtain the final microenvironment parameters.

[0012] Thirdly, an electronic device is provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements any of the methods of the first aspect.

[0013] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements any of the methods of the first aspect.

[0014] This invention provides a method, device, electronic equipment, and medium for monitoring the microenvironment of a photovoltaic (PV) power station. The PV power station is divided into an under-panel area, an inter-panel area, and an external array area. Different types of sensor groups are deployed in each area. Initial microenvironmental parameters of the PV power station are acquired from the sensor groups in each area. Using the initial microenvironmental parameters of the external array area as constraints, a preset correction algorithm is used to correct the initial microenvironmental parameters of the under-panel and inter-panel areas, yielding the final microenvironmental parameters. This invention achieves logical verification of the complex internal microenvironment by introducing macroscopic data from outside the array as constraints. It effectively eliminates false high-temperature or high-wind readings caused by localized strong radiation, dust accumulation, or sensor drift, significantly improving the accuracy and robustness of the monitoring data.

[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of the microenvironment monitoring method for photovoltaic power stations provided in an embodiment of the present invention is shown; Figure 2 A flowchart of another method for monitoring the microenvironment of a photovoltaic power station provided by an embodiment of the present invention is shown; Figure 3 A schematic diagram of the structure of a microenvironment monitoring device for a photovoltaic power station provided in an embodiment of the present invention is shown; Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present invention is shown. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0019] Given the significant limitations of existing photovoltaic (PV) power plant monitoring technologies, such as the fact that large ground-mounted power plants often only have a standard weather station (collecting only ambient temperature, wind speed, and irradiance) at the outer edge of the array, measurement errors can occur when sensors inside the PV array drift due to localized strong heat radiation, dust accumulation, or airflow turbulence. This reduces the accuracy and reliability of the monitoring data, making it difficult to meet the needs of refined operation and maintenance and efficient power generation in PV power plants.

[0020] Based on this, embodiments of the present invention provide a method and apparatus for monitoring the microenvironment of a photovoltaic power station, which are described below through embodiments.

[0021] This invention provides a micro-environment monitoring method for photovoltaic power plants. The core of this method lies in constructing a logical framework of "zonal monitoring - boundary constraints - global correction" to solve the problem that local data distortion is difficult to detect in traditional monitoring.

[0022] The method first divides the photovoltaic power station into the area under the panels, the area between the panels, and the area outside the array; for each of these three areas, sensor groups adapted to their environmental characteristics are deployed.

[0023] Specifically, this classification is based on the physical disturbance mechanism of the photovoltaic array on the microenvironment.

[0024] The area under the photovoltaic module refers to the area directly below the module that is affected by shading. The light intensity in this area is reduced and the humidity is often high. Soil temperature and humidity sensors, air temperature and humidity sensors, and illuminance sensors are usually deployed in this area.

[0025] The inter-panel area refers to the channel area between two adjacent rows of photovoltaic modules. Due to the arrangement of the modules, the airflow and temperature fields in this area have special distribution characteristics, and wind speed and direction meters, infrared thermal imagers, etc. are usually deployed there.

[0026] The area outside the array refers to the open area around the photovoltaic array that is not shaded by the components. This area is directly exposed to the natural atmospheric environment and is usually equipped with a standard weather station, including a high-precision thermometer, hygrometer, anemometer, and irradiance sensor.

[0027] It should be understood that the above sensor types are only examples, and can be flexibly adjusted according to monitoring needs during actual deployment, such as adding sand and dust sensors in a solar-sand hybrid power station.

[0028] like Figure 1 As shown, the method includes: Step S101: Obtain the initial microenvironmental parameters of the photovoltaic power station collected by the sensor groups in each region.

[0029] Specifically, raw data from the photovoltaic power station is collected in real time through anemometers and temperature sensors deployed under the panels, infrared thermal imagers or contact temperature probes deployed between the panels, and standard weather stations (including thermometers, hygrometers, anemometers, and irradiance sensors) located outside the array. This raw data is referred to as initial microenvironmental parameters, including temperature (…). ), wind speed ( ), relative humidity ( ) and solar irradiance ( Physical quantities such as ).

[0030] Step S102: Using the initial microenvironment parameters of the outer region of the array as constraints, the initial microenvironment parameters of the under-board region and the inter-board region are corrected using a preset correction algorithm to obtain the final microenvironment parameters.

[0031] Specifically, the data from the region outside the array represents the macro-meteorological boundary conditions at the location of the site. Its values ​​are not significantly affected by the heat island effect of photovoltaic modules and local airflow disturbances, and therefore can be used as a true benchmark or prior information.

[0032] The correction algorithm utilizes this physical characteristic to establish a correlation constraint between the macroscopic environment and the internal micro-environment. For example, when the wind speed outside the array changes abruptly, the wind speed in the areas below and between the panels should theoretically show a corresponding lag or attenuation. If the readings of the sensors below the panels deviate significantly from the trend outside the array without a reasonable physical explanation, it is determined to be a measurement deviation and corrected.

[0033] This embodiment achieves logical verification of the complex internal microenvironment by introducing macroscopic data from outside the array as constraints. This method effectively eliminates false high-temperature or high-wind readings caused by localized strong radiation, dust accumulation, or sensor drift, significantly improving the accuracy and robustness of the monitoring data.

[0034] Initial microenvironment parameters often contain outliers caused by noise, packet loss, or interference from local heat island effects. For example, sensors in the under-board area may have inflated readings due to component reflections, or measurement deviations may occur due to dust accumulation. If this raw data is used directly, it cannot accurately reflect the true microenvironment state.

[0035] Therefore, based on the above embodiments, before correcting the initial microenvironment parameters, the method further includes: Step S103: Perform outlier removal processing on the initial microenvironment parameters of each region to obtain the first data.

[0036] Due to the complex environment at photovoltaic sites, sensors may generate outliers due to electromagnetic interference, equipment failure, or extreme weather.

[0037] This embodiment preferably uses the 3σ criterion for elimination, that is, assuming that normal data follows a Gaussian distribution, if the temperature value collected at a certain moment... Exceeding the dynamic threshold window If the value is an outlier, it will be removed. and These are the mean and standard deviation of the historical data, respectively.

[0038] It should be understood that, in addition to statistical methods, machine learning algorithms such as isolated forests can also be used for anomaly detection for non-Gaussian distributed data features.

[0039] The dataset after removing outliers is the first dataset, effectively avoiding misleading the system's judgment by false alarm signals.

[0040] Step S104: Use linear interpolation to complete the first data to obtain the second data.

[0041] Outlier removal or signal loss can cause breakpoints in the time series. If incomplete data is used directly, the subsequent Kalman filter algorithm may fail to converge.

[0042] Therefore, this embodiment uses linear interpolation to fill in the missing time points. numerical value The formula is as follows:

[0043] in, and These are the two most recent valid moments before and after the missing moment. and These are the corresponding observations. For short-term missing values ​​(e.g., less than 5 minutes), linear interpolation can better approximate the continuous trend of physical quantities, ensuring the integrity of the time series.

[0044] Step S105: Normalize the second data and map it to a unified interval to obtain the preprocessed initial microenvironment parameters.

[0045] Because the dimensions and orders of magnitude of different physical quantities differ greatly (e.g., temperature is in degrees Celsius, while the order of magnitude is in...), The irradiance is watts per square meter, reaching levels up to [missing information]. Directly inputting the correction algorithm will cause large values ​​to dominate the calculation process, masking the influence of small value features.

[0046] Therefore, this embodiment uses the Min-Max normalization method, as shown in the following formula:

[0047] in, These are the original values. and These are the minimum and maximum values ​​of the physical quantity within a specific time period. The normalized value ranges from 1 to 10. between.

[0048] Normalization eliminates dimensional differences, unifies data scale, provides standardized input for subsequent Bayesian estimation and Kalman filtering algorithms, significantly reduces the difficulty of algorithm convergence, and improves the overall system stability.

[0049] Based on the above embodiments, the initial microenvironment parameters of the outer region of the array are used as constraints. A preset correction algorithm is used to correct the initial microenvironment parameters of the under-board region and the inter-board region, resulting in the following final microenvironment parameters: Step S102A: Perform internal fusion of the initial microenvironment parameters of each region to obtain single fused data for each region.

[0050] Since multiple sensors of the same type are often deployed in the same area, the data collected by these sensors is redundant. To improve the representativeness of the data, this embodiment adopts a weighted fusion strategy.

[0051] Taking the area under the plate as an example, assuming N temperature sensors are deployed, the reading of the i-th sensor is... The corresponding confidence weight is (Weights can be determined by sensor accuracy level or historical stability data), then the single fused data for this region The calculation formula is:

[0052] Similarly, the above fusion operation is performed on the inter-board region and the region outside the array respectively to obtain a single fused data vector for each of the three regions.

[0053] This step effectively reduces the random fluctuations of single-point measurements and extracts the principal values ​​of environmental features for each region.

[0054] Step S102B: Using the single fused data of the outer region of the array as a priori, the Bayesian estimation method is used to correct the spatial error of the single fused data of the under-panel and inter-panel regions to obtain the intermediate microenvironment parameters of the three regions.

[0055] The core idea of ​​Bayesian estimation is to use known information to update the probability distribution of unknown states.

[0056] In this embodiment, the region outside the array is directly exposed to the natural atmospheric environment, and its data represents the macroscopic meteorological boundary conditions of the site location. It is not significantly affected by the photovoltaic module heat island effect and local airflow disturbances, and therefore is regarded as prior information about the real environmental conditions. The fused data of the under-panel and inter-panel regions, on the other hand, are observations affected by local disturbances.

[0057] Assuming the actual microenvironment state is X, and the observation model is... Where V is the observation noise, The observation matrix is ​​given. The prior distribution provided by the out-of-array data is... ,in The predicted state is derived from data outside the array. The prior covariance. According to Bayes' theorem, the posterior estimate is... The calculation is as follows:

[0058] Where K is the Kalman gain matrix, its calculation formula is:

[0059] Where R is the observation noise covariance matrix.

[0060] This formula is used to smoothly correct the observed values ​​in the under-plate and inter-plate regions to the prior values ​​outside the array. For example, if a sensor under the plate has an excessively high reading due to the local heat island effect, while the temperature outside the array is normal, Bayesian estimation will pull the reading back to the normal range based on the gain matrix K, thereby eliminating the systematic bias of the local sensor.

[0061] The corrected data are the intermediate microenvironment parameters, which retain the local characteristics of the subplate and interplate regions while conforming to macroscopic meteorological and physical laws.

[0062] Step S102C: Based on the Kalman filter algorithm, the dynamic disturbance of the intermediate microenvironment parameters over time is corrected to obtain the final microenvironment parameters.

[0063] Although the intermediate microenvironment parameters, after spatial error correction, eliminate systematic biases, random noise may still exist in the time dimension, and the microenvironment state itself has the physical characteristic of continuous evolution. Therefore, this embodiment introduces the Kalman filter algorithm to further filter out high-frequency noise by utilizing the correlation on the time series, restoring the true dynamic change trend. The specific process will be elaborated in detail in subsequent embodiments.

[0064] This embodiment achieves spatial dimension data fusion and error correction through Bayesian estimation, effectively solving the problem of low reliability of sensor data in a single region. Kalman filtering removes noise in the time series, providing high-quality input for subsequent power plant efficiency assessments.

[0065] Based on the above embodiments, this embodiment provides a detailed description of the time-series dynamic correction process of microenvironment parameters.

[0066] The core logic of the Kalman filter algorithm lies in its iterative "prediction-update" mechanism. This algorithm assumes that changes in the microenvironment follow certain physical laws, and that the observations contain noise. By continuously fusing predictions based on a physical model with sensor-based observations, it achieves an optimal estimate of the true state.

[0067] Specifically, based on the Kalman filter algorithm, the dynamic perturbations of the intermediate microenvironment parameters over time are corrected to obtain the final microenvironment parameters, including: Step S102C1: Using the intermediate microenvironment parameters from the previous moment as input, calculate the predicted parameters for the current moment based on the pre-built state-space model.

[0068] This step is state prediction. It involves estimating the prior state of the microenvironment parameters at the current moment. It can be calculated using the state prediction equation:

[0069] In the formula, This represents the optimal estimate of the intermediate microenvironment parameters at time k-1; This is the state transition matrix; To control the input matrix; For external control inputs (such as the rate of change of light intensity and the rate of change of external wind speed).

[0070] Wherein, the state transition matrix It is the core parameter of the Kalman filter, which determines how the state evolves over time.

[0071] In this embodiment, It is not a simple empirical coefficient, but is constructed based on physical mechanisms. The specific construction process will be described in detail in subsequent embodiments.

[0072] Using this equation, the system leverages historical information from the previous moment, combined with a physical model, to calculate the theoretical prediction value for the current moment. This process demonstrates the ability to predict the evolutionary trend of the microenvironment.

[0073] Step S102C2: Based on the deviation between the predicted parameters and the actual observation parameters of the sensor group at the current time, the predicted parameters are corrected to obtain the final microenvironment parameters.

[0074] This step involves updating the observations. Since the predicted values ​​are based solely on model calculations, errors are inevitable, and the sensor observations... It also includes measurement noise, so it needs to be corrected for the deviation between the two.

[0075] First, calculate the Kalman gain. The calculation formula is as follows:

[0076] in, For the prediction error covariance matrix, For the observation matrix, To observe the noise covariance matrix.

[0077] Kalman gain It characterizes the trust weight between predicted and observed values: when observation noise... When smaller, As the model's prediction error increases, the system places greater trust in the observed values; the system also tends to accept the observed values ​​more readily when the model's prediction error is large. The final posterior state estimate... (i.e., the final microenvironment parameters) are calculated through the update equation:

[0078] in, The term "new information" or "residual" reflects the deviation between actual observations and theoretical predictions.

[0079] The Kalman filter algorithm dynamically corrects the predicted value by multiplying this bias by a gain coefficient and feeding it back into the state estimate. For example, if the predicted temperature under the plate will rise slowly, but the actual observation suddenly shows a spike (possibly electromagnetic interference), the Kalman filter, having established a smooth prediction trend, and with the observation noise covariance... It will adaptively adjust based on historical data, at which point the gain will... It will automatically reduce, thereby effectively filtering out the random noise and keeping the final output temperature curve smooth and close to the real physical process.

[0080] This embodiment further improves the smoothness and tracking accuracy of microenvironment parameters through timing correction using Kalman filtering, ensuring the stability of the output data.

[0081] State transition matrix (Or denoted as matrix A) is the core parameter that determines the accuracy of the prediction. If this matrix is ​​obtained only through data fitting and lacks physical constraints, the prediction results will often deviate significantly from the actual situation when encountering extreme weather or sensor failure outside the training samples.

[0082] Therefore, this embodiment proposes a state-space model construction method based on physical mechanisms, which explicitly encodes the thermodynamic and fluid dynamic laws of photovoltaic power stations into the algorithm model.

[0083] Specifically, the process of constructing a state-space model includes: Step A: Based on the laws of energy conservation, mass conservation, and momentum conservation, establish a set of global heat flow coupled evolution equations characterizing regional heat and mass transfer and fluid motion.

[0084] The evolution of the microenvironment of a photovoltaic power station is not a random process, but strictly follows the laws of physics.

[0085] The energy conservation equation describes the spatiotemporal distribution of the temperature field, taking into account solar radiation input, air convection heat transfer, and radiation heat dissipation from the component surface.

[0086] The energy conservation equation is as follows:

[0087] in, air density, For temperature; For time; For specific heat capacity, Thermal conductivity, This is the solar radiation absorption term. For convective heat transfer, This is a radiative heat transfer term.

[0088] The mass conservation equation describes the continuity of airflow in a flow field, ensuring the balance of air mass entering and exiting each region. The mass conservation equation is as follows:

[0089] in, This represents the instantaneous velocity of the airflow; other parameters are the same as above.

[0090] The momentum conservation equation (Navier-Stokes equation) describes the pressure distribution and velocity evolution under the obstruction of the photovoltaic array, considering wind speed and direction. The formula is as follows:

[0091] in, For pressure, For dynamic viscosity, This refers to volumetric forces (such as buoyancy), and other parameters are the same as above.

[0092] These three equations together form a set of global heat flow coupled evolution equations.

[0093] Step B: Divide the photovoltaic array into several grid cells, and use the finite volume method to discretize the global heat flow coupling evolution equations on the grid cells to obtain the state evolution equations of each grid cell.

[0094] Since analytical solutions are difficult to obtain, this embodiment employs a numerical method. The three-dimensional space encompassing the area beneath the plates, between the plates, and outside the array is divided into a finite number of grid cells. Using the finite volume method, the control volume of each grid cell is integrated, transforming the continuous partial differential equations into algebraic equations.

[0095] For example, for the j-th grid cell, its energy conservation equation can be discretized as:

[0096] in, For unit volume, For time step; The temperature at the current moment; The temperature at the next moment; The heat transfer admittance between adjacent units; This represents the convective heat transfer term; For source terms (such as solar energy input).

[0097] This step decomposes complex continuous physical fields into a computable network of discrete nodes.

[0098] Step C: Analyze the state evolution equations to extract the thermal coupling coefficient, momentum transfer operator, and mass transfer operator that characterize the interaction between adjacent units.

[0099] The thermal coupling coefficient reflects the ability of adjacent grids to exchange heat, corresponding to the equation in... item.

[0100] The momentum transport operator reflects the diffusion and convection of fluid momentum between grids, corresponding to the viscous and pressure gradient terms in the Navier-Stokes equations.

[0101] The mass transport operator reflects the flux of airflow across the grid boundary.

[0102] This step analyzes the equation structure to quantify these physical interactions into specific operator parameters, thereby capturing the interconnected relationships where a change in one part affects the whole. For example, changes in wind speed under the plates directly influence the pressure field between the plates through the momentum operator, thus altering the heat dissipation efficiency.

[0103] Step D: Map the thermal coupling coefficient, momentum transfer operator, and mass transfer operator to their corresponding positions in the preset state transition matrix framework to generate the state transition matrix, and use the state transition matrix to construct the state space model.

[0104] Assemble the extracted physical operators into a matrix form. Define the state vector. If the state transition matrix includes variables such as temperature and wind speed of all grid cells, then... elements in That is, it is determined by the coupling operator of the j-th grid to the i-th grid.

[0105] If two grids are not adjacent in space or have no direct physical interaction, the matrix element at the corresponding position is zero.

[0106] The resulting matrix Its sparse structure and numerical magnitude are entirely determined by physical topology and material properties, rather than by black-box fitting. This means that even in blind spots where there is no sensor data, the model can still deduce reasonable evolutionary trends based on physical laws.

[0107] Specifically, based on the state transition matrix Constructing a state-space model:

[0108] in, This is the state transition matrix, whose elements Representing the The state of the i-th grid is related to the i-th The degree of influence of each grid (determined by thermal coupling coefficient, etc.) B For the input matrix, External stimuli (such as wind speed outside the array, irradiance). This is process noise.

[0109] This embodiment constructs a state-space model with strong physical interpretability by explicitly embedding physical conservation laws into the state transition matrix. This model not only improves the prediction accuracy of Kalman filtering, but more importantly, enhances the robustness of the system.

[0110] When sensors fail due to malfunctions or encounter extreme conditions such as sudden gusts of wind or cloud cover, purely data-driven models often fail. However, the physical model constructed in this embodiment can still provide predictive values ​​that conform to physical logic based on the laws of conservation of energy and momentum, thus avoiding the collapse of the monitoring system and greatly improving the engineering practicality of the microenvironment monitoring solution.

[0111] Based on the above embodiments, sensor groups are deployed in each region according to a preset sensor layout scheme, such as... Figure 2 As shown, the process of determining the preset sensor layout scheme includes: Step S106: Obtain initial microenvironmental rough data of the photovoltaic power station site.

[0112] In the initial stages of power plant construction or during the operation and maintenance assessment phase, it is essential to first understand the environmental baseline of the site. This step utilizes mobile weather monitoring vehicles, portable sensors mounted on drones, or temporarily deployed low-density sensor networks to conduct rapid scanning or sparse sampling of the entire site. The collected data includes spatial distribution snapshots of parameters such as temperature, humidity, and wind speed; this data is referred to as initial microenvironmental coarse-grained data.

[0113] It should be understood that the data accuracy requirements at this stage are relatively low, with the focus on the global coverage to provide a data foundation for subsequent spatial analysis.

[0114] Step S107: Perform spatial variation analysis on the initial coarse-collected microenvironment data to extract spatially relevant distance parameters.

[0115] The spatial distribution of microenvironment parameters often exhibits continuity and correlation; points that are closer together tend to have more similar parameter values. To quantify this spatial correlation, this embodiment introduces the semivariogram from geostatistics for analysis. The semivariogram describes the relationship between the differences between two points in space as distance increases. The specific formula is as follows:

[0116] in, For position Environmental parameter values ​​at the location; The spatial distance between two points; The distance interval is The number of point pairs.

[0117] By calculating the semivariogram values ​​at different distances and fitting the curves, the key parameter can be extracted: the spatially relevant distance parameter (range).

[0118] The spatial correlation distance parameter is the distance threshold at which the fitted curve reaches the sill value, representing the maximum effective range within which microenvironmental parameters have a significant spatial correlation. Beyond this distance, the correlation between two points weakens rapidly, and environmental characteristics tend to become independent. This parameter intuitively reflects the radius of influence of microenvironmental changes.

[0119] Step S108: Based on the preset correspondence between spatially related distance parameters and parameter variation degree, identify homogeneous regions and heterogeneous mutation zones of parameter variation.

[0120] Based on the extracted spatial correlation distance parameters and local variation coefficients, the system classifies the field area into regional types.

[0121] The homogeneous region refers to a continuous area where the variation range of microenvironmental parameters is less than a preset threshold and the spatial correlation meets the spatial correlation distance parameter. Such regions are typically located in open areas at the center of the site, where environmental changes are gradual and gradients are small.

[0122] Heterogeneous abrupt change zones refer to transitional regions where microenvironmental parameters undergo drastic changes or where spatial correlation rapidly decays to zero over short distances. These regions are typically located at the edges of photovoltaic arrays, under-panel air duct inlets, shadow boundaries, or abrupt topographic changes, characterized by dramatic environmental shifts and large gradients.

[0123] This identification process simplifies the complex site space into several distinctive sub-regions, providing a basis for differentiated deployment.

[0124] Step S109: Based on the identified homogeneous regions and heterogeneous transition zones, adjust the sensor deployment density in each region to obtain an adaptive sensor layout scheme for each region.

[0125] This embodiment employs drastically different deployment strategies for different types of regions.

[0126] In homogeneous regions, since environmental parameters change slowly and continuously, the sensor spacing is set to be a multiple of the spatially relevant distance parameters (e.g., 1.5 or 2 times) to reduce redundant monitoring and lower hardware costs.

[0127] In heterogeneous transition zones, due to the drastic changes in environmental parameters and weak spatial correlation, the sensor spacing is set to be less than the spatial correlation distance parameter (e.g., 0.5 times), or even deployed in a denser manner, to ensure that rapidly changing environmental details can be captured and to avoid monitoring blind spots.

[0128] The final layout scheme is a non-uniform, adaptive sensor network topology that achieves an optimal balance between monitoring accuracy and economic cost.

[0129] This embodiment uses spatial variation analysis to accurately deploy limited sensor resources to key areas where environmental changes are most drastic, avoiding the waste of resources or omission of key information caused by the traditional uniform distribution method, and significantly improving the overall efficiency of the monitoring system.

[0130] The microenvironment of a photovoltaic power station is not static. With seasonal changes, vegetation growth, equipment aging, or changes in the surrounding terrain (such as the construction of new buildings), what was originally a homogeneous area may evolve into a heterogeneous zone, or the data quality of the originally deployed sensors may deteriorate due to malfunctions. If the fixed layout scheme from the initial construction phase is still used, it may lead to monitoring blind spots or data distortion. Therefore, this embodiment introduces a closed-loop feedback mechanism to achieve dynamic optimization of the layout scheme.

[0131] Specifically, the method also includes: Step S110: During the operation of the photovoltaic power station, monitor the variance of the microenvironmental parameters of each zone in real time.

[0132] The system continuously calculates the statistical variance of microenvironmental parameters in each region (underboard, betweenboards, and outside the array) within a sliding time window in the background. Variance is a statistic that measures the degree of data fluctuation, and its calculation formula is as follows:

[0133] in, For a moment The monitoring value, The average value within the window. The variance represents the window length. An increase in variance indicates greater volatility in the microenvironment of the region, potentially indicating the presence of new heterogeneous abrupt change zones or sensor anomalies.

[0134] Step S111: If the variance of any region exceeds a preset variance threshold for a preset number of consecutive preset times, the sensor layout scheme determination process is re-triggered, and sensor movement or supplementation suggestions are generated.

[0135] To avoid false triggering caused by momentary interference, this embodiment sets up an anti-jitter mechanism for a preset number of consecutive cycles.

[0136] For example, setting a preset variance threshold is The preset number of times is Next. When the variance of a certain region consecutive occurrences After one sampling period, it was determined that the environmental characteristics of the area had undergone a fundamental change.

[0137] At this point, the spatial variability analysis process is automatically re-executed: coarse data for the region is re-collected, a new semivariogram is calculated, and new homogeneous regions and heterogeneous abrupt change zones are identified. If the variability of an existing homogeneous region is found to have intensified, transforming into a heterogeneous abrupt change zone, the system will generate supplementary suggestions, prompting maintenance personnel to add sensors at specific coordinate points for more intensive monitoring. If a region is found to be in a state of high variance for a long period and sensor readings are abnormal, relocation or maintenance suggestions will be generated, prompting the relocation of sensors in redundant areas to hotspot areas or the replacement of faulty sensors.

[0138] This embodiment enables the sensor layout scheme to evolve dynamically with changes in the environment, ensuring that the monitoring system maintains optimal monitoring performance throughout its entire life cycle, thus solving the problem of poor adaptability of traditional fixed layout schemes.

[0139] Based on the same inventive concept, a microenvironment monitoring device for a photovoltaic power station is provided. The photovoltaic power station is divided into an under-panel area, an inter-panel area, and an external array area; each area is equipped with different types of sensor groups; such as... Figure 3 As shown, the device includes: Acquisition unit 301 is used to acquire the initial microenvironmental parameters of the photovoltaic power station collected by the sensor groups in each region; The correction unit 302 is used to use the initial microenvironment parameters of the outer region of the array as constraints, and to use a preset correction algorithm to correct the initial microenvironment parameters of the under-board region and the inter-board region to obtain the final microenvironment parameters.

[0140] Based on the same technical concept, embodiments of the present invention also provide an electronic device, such as... Figure 4 As shown, it includes a processor 401, a communication interface 402, a memory 403, and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404.

[0141] Memory 403 is used to store computer programs; When the processor 401 executes the program stored in the memory 403, it implements the steps of the above method embodiment.

[0142] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0143] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0144] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0145] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0146] In another embodiment of the present invention, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described method embodiments. Specific implementation details can be found in the method embodiments and will not be repeated here.

[0147] The microenvironment monitoring device for photovoltaic power plants provided in this embodiment of the invention can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in this embodiment of the invention are the same as those in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiments can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the above method embodiments, and will not be repeated here.

[0148] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0149] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0150] In addition, the functional units in the embodiments provided by the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0151] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0152] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0153] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. All should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method of micro-environment monitoring of a photovoltaic plant, characterized in that, The photovoltaic power station is divided into an under-panel area, an inter-panel area, and an external array area; different types of sensor groups are deployed in each area; the method includes: Acquire the initial microenvironmental parameters of the photovoltaic power station collected by sensor groups in each region; Using the initial microenvironment parameters of the outer region of the array as constraints, the initial microenvironment parameters of the under-board region and the inter-board region are corrected using a preset correction algorithm to obtain the final microenvironment parameters.

2. The method of claim 1, wherein, Before correcting the initial microenvironment parameters, the method further includes: Outlier removal was performed on the initial microenvironment parameters of each region to obtain the first data. The first data is completed using linear interpolation to obtain the second data; The second data is normalized and mapped to a unified interval to obtain the preprocessed initial microenvironment parameters.

3. The method according to claim 1, characterized in that, The initial microenvironment parameters of the outer region of the array are used as constraints, and a preset correction algorithm is used to correct the initial microenvironment parameters of the under-board region and the inter-board region to obtain the final microenvironment parameters, including: The initial microenvironment parameters of each region are internally fused to obtain single fused data for each region. Using the single fused data of the region outside the array as a priori, the Bayesian estimation method is used to correct the spatial error of the single fused data of the region under the board and between the boards, so as to obtain the intermediate microenvironment parameters of the three regions. Based on the Kalman filter algorithm, the dynamic perturbation of the intermediate microenvironment parameters over time is corrected to obtain the final microenvironment parameters.

4. The method according to claim 3, characterized in that, The Kalman filter algorithm is used to correct the dynamic perturbations of the intermediate microenvironment parameters over time, resulting in the final microenvironment parameters, including: Using the intermediate microenvironment parameters from the previous moment as input, the predicted parameters for the current moment are calculated based on the pre-built state-space model; Based on the deviation between the predicted parameters and the actual observation parameters of the sensor group at the current time, the predicted parameters are corrected to obtain the final microenvironment parameters.

5. The method according to claim 4, characterized in that, The process of constructing the state-space model includes: Based on the laws of energy conservation, mass conservation, and momentum conservation, a set of global heat flow coupled evolution equations characterizing regional heat and mass transfer and fluid motion is established. The photovoltaic array is divided into several grid cells, and the global heat flow coupling evolution equations are discretized on the grid cells using the finite volume method to obtain the state evolution equations of each grid cell. The state evolution equations are analyzed to extract the thermal coupling coefficient, momentum transfer operator, and mass transfer operator that characterize the interaction between adjacent units; The thermal coupling coefficient, momentum transfer operator, and mass transfer operator are mapped to their corresponding positions in a preset state transition matrix framework to generate a state transition matrix, and a state space model is constructed using the state transition matrix.

6. The method according to claim 1, characterized in that, Sensor groups are deployed in each area according to a preset sensor layout scheme. The process of determining the preset sensor layout scheme includes: Obtain initial rough data of the microenvironment of the photovoltaic power plant site; Spatial variation analysis was performed on the initial coarse-collected microenvironment data to extract spatially relevant distance parameters; Based on the pre-defined correspondence between spatially relevant distance parameters and the degree of parameter variation, homogeneous regions and heterogeneous mutation zones of parameter variation are identified. Based on the identified homogeneous regions and heterogeneous transition zones, the sensor deployment density in each region is adjusted to obtain an adaptive sensor layout scheme for each region.

7. The method according to claim 5, characterized in that, The method further includes: During the operation of the photovoltaic power station, the variance of the microenvironment parameters in each zone is monitored in real time. If the variance of any region exceeds a preset variance threshold for a preset number of consecutive preset times, the sensor layout determination process will be retried, and suggestions for sensor movement or addition will be generated.

8. A microenvironment monitoring device for a photovoltaic power station, characterized in that, The photovoltaic power station is divided into an under-panel area, an inter-panel area, and an external array area; each area is equipped with different types of sensor groups; the device includes: The acquisition unit is used to acquire the initial microenvironmental parameters of the photovoltaic power station collected by the sensor groups in each area; The calibration unit is used to use the initial microenvironment parameters of the outer region of the array as constraints, and to use a preset calibration algorithm to calibrate the initial microenvironment parameters of the under-board region and the inter-board region to obtain the final microenvironment parameters.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes the program stored in the memory, it implements the method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images